Integrating copy number polymorphisms into array CGH analysis using a robust HMM

Abstract

Source code written in Matlab is available from http://www.cs.ubc.ca/~sshah/acgh.

Genome-wide profiling of follicular lymphoma by array comparative genomic hybridization reveals prognostically significant DNA copy number imbalances

“…), using a criterion for an aberration defined as an apparent log-ratio shift away from baseline in a minimum of 3 adjacent BACs (ϳ200 kb or larger), and computational analysis by determining probability of aberration (loss, neutral, or gain) for each clone using the program CNA-HMMer version 0.1 (available at http://www.cs.ubc.ca/ϳsshah/acgh/), which is based on a Hidden Markov Model (HMM). 37 Only those alterations identified by both HMM and visual interpretation were accepted as true. We modified the emission model of the HMM described in Shah et al to be a mixture of Student t distributions, achieving the equivalent robustness to outliers while producing output that was more interpretable to the investigator.…”

“…We modified the emission model of the HMM described in Shah et al to be a mixture of Student t distributions, achieving the equivalent robustness to outliers while producing output that was more interpretable to the investigator. 37 In addition, this modification required fewer hyperparameters to be set, which were selected automatically using an 'empirical Bayes'-type approach. 38 Concordance between the visual calls and the HMM predictions was assessed by calculating the area under the receiver operator characteristic (ROC) curve.…”

Array-comparative genomic hybridization in sporadic benign pheochromocytomas

“…Copy number alterations were identified via data visualization using custom software called 'SeeGH' (freely available at http://www.flintbox.ca/technology. asp?techZFB312FB) and loss, normal, and gain probabilities for each clone as determined by a modified hidden Markov model , Shah et al 2006. Data were filtered based on both replicate S.D.…”

Getting it right: designing microarray (and not ‘microawry') comparative genomic hybridization studies for cancer research

“…With the anticipated introduction of a 1000K SNP array later this year, the effective resolution of OaCGH arrays may increase to as much as B25 kb. Alternatively, the data need to be analysed using algorithms that reduce the experimental variation for regions with similar copy numbers (ie smoothing algorithms), such as adaptive weighted smoothing (aws), 71 maximum likelihood models, hidden Markov models, 72 or row Loess methods 73 and Gaussian smoothing, 74 as per BAC array data, before confidently defining genomic changes. In simplistic terms, these analytical methods transform aCGH data by organizing a user-defined consecutive sequence of adjacent signals into regions of constant copy number known as segments, which are subsequently classified as a gain, a loss, or no change depending on their signal intensities.…”

“…8,89,90 Alternatively, as discussed earlier, segmentation algorithms can be used to define the boundaries of copy number changes. [71][72][73] We have successfully used a combination of the two approaches and identified copy number changes that can be CISH/FISH verified. 8,9,15,91,92 Whichever methods utilized in the analysis of aCGH data, it is vitally important that the invariably and often excessively large volume of data generated is appropriately curated and validated with in situ or molecular methods.…”